Sums of products of polynomials in few variables : lower bounds and polynomial identity testing

We study the complexity of representing polynomials as a sum of products of polynomials in few variables. More precisely, we study representations of the form $$P = \sum_{i = 1}^T \prod_{j = 1}^d Q_{ij}$$ such that each $Q_{ij}$ is an arbitrary polynomial that depends on at most $s$ variables. We prove the following results. 1. Over fields of characteristic zero, for every constant $\mu$ such that $0 \leq \mu < 1$, we give an explicit family of polynomials $\{P_{N}\}$, where $P_{N}$ is of degree $n$ in $N = n^{O(1)}$ variables, such that any representation of the above type for $P_{N}$ with $s = N^{\mu}$ requires $Td \geq n^{\Omega(\sqrt{n})}$. This strengthens a recent result of Kayal and Saha [KS14a] which showed similar lower bounds for the model of sums of products of linear forms in few variables. It is known that any asymptotic improvement in the exponent of the lower bounds (even for $s = \sqrt{n}$) would separate VP and VNP[KS14a]. 2. We obtain a deterministic subexponential time blackbox polynomial identity testing (PIT) algorithm for circuits computed by the above model when $T$ and the individual degree of each variable in $P$ are at most $\log^{O(1)} N$ and $s \leq N^{\mu}$ for any constant $\mu < 1/2$. We get quasipolynomial running time when $s < \log^{O(1)} N$. The PIT algorithm is obtained by combining our lower bounds with the hardness-randomness tradeoffs developed in [DSY09, KI04]. To the best of our knowledge, this is the first nontrivial PIT algorithm for this model (even for the case $s=2$), and the first nontrivial PIT algorithm obtained from lower bounds for small depth circuits.

[1]  Nitin Saxena,et al.  Jacobian hits circuits: hitting-sets, lower bounds for depth-D occur-k formulas & depth-3 transcendence degree-k circuits , 2011, STOC '12.

[2]  Avi Wigderson,et al.  Depth-3 arithmetic circuits over fields of characteristic zero , 2002, computational complexity.

[3]  Richard Zippel,et al.  Probabilistic algorithms for sparse polynomials , 1979, EUROSAM.

[4]  Neeraj Kayal,et al.  A super-polynomial lower bound for regular arithmetic formulas , 2014, STOC.

[5]  Partha Mukhopadhyay,et al.  Depth-4 Identity Testing and Noether's Normalization Lemma , 2016, CSR.

[6]  Oleg Golberg Combinatorial Nullstellensatz , 2007 .

[7]  Nitin Saxena,et al.  Quasi-polynomial hitting-set for set-depth-Δ formulas , 2012, STOC '13.

[8]  Amir Shpilka,et al.  Subexponential Size Hitting Sets for Bounded Depth Multilinear Formulas , 2016, computational complexity.

[9]  Ran Raz,et al.  Separation of Multilinear Circuit and Formula Size , 2006, Theory Comput..

[10]  Ankit Gupta Algebraic Geometric Techniques for Depth-4 PIT & Sylvester-Gallai Conjectures for Varieties , 2014, Electron. Colloquium Comput. Complex..

[11]  Amir Shpilka Affine projections of symmetric polynomials , 2001, Proceedings 16th Annual IEEE Conference on Computational Complexity.

[12]  Shubhangi Saraf,et al.  On the Power of Homogeneous Depth 4 Arithmetic Circuits , 2014, 2014 IEEE 55th Annual Symposium on Foundations of Computer Science.

[13]  Leslie G. Valiant,et al.  Completeness classes in algebra , 1979, STOC.

[14]  Michael A. Forbes Deterministic Divisibility Testing via Shifted Partial Derivatives , 2015, 2015 IEEE 56th Annual Symposium on Foundations of Computer Science.

[15]  N. Nisan Lower Bounds for Non-Commutative Computation (Extended Abstract) , 1991, STOC 1991.

[16]  Shubhangi Saraf,et al.  The limits of depth reduction for arithmetic formulas: it's all about the top fan-in , 2013, Electron. Colloquium Comput. Complex..

[17]  Ilya Volkovich,et al.  Black-Box Identity Testing of Depth-4 Multilinear Circuits , 2011, Combinatorica.

[18]  Amir Shpilka,et al.  On identity testing of tensors, low-rank recovery and compressed sensing , 2011, STOC '12.

[19]  V. Vinay,et al.  Arithmetic Circuits: A Chasm at Depth Four , 2008, 2008 49th Annual IEEE Symposium on Foundations of Computer Science.

[20]  Neeraj Kayal,et al.  Arithmetic Circuits: A Chasm at Depth Three , 2013, 2013 IEEE 54th Annual Symposium on Foundations of Computer Science.

[21]  Zeev Dvir,et al.  Hardness-randomness tradeoffs for bounded depth arithmetic circuits , 2008, SIAM J. Comput..

[22]  Neeraj Kayal,et al.  Approaching the Chasm at Depth Four , 2013, 2013 IEEE Conference on Computational Complexity.

[23]  Nutan Limaye,et al.  An Exponential Lower Bound for Homogeneous Depth Four Arithmetic Formulas , 2014, 2014 IEEE 55th Annual Symposium on Foundations of Computer Science.

[24]  Neeraj Kayal,et al.  Lower Bounds for Sums of Products of Low arity Polynomials , 2015, Electron. Colloquium Comput. Complex..

[25]  Noam Nisan,et al.  Lower bounds for non-commutative computation , 1991, STOC '91.

[26]  Neeraj Kayal,et al.  Lower Bounds for Depth-Three Arithmetic Circuits with small bottom fanin , 2016, computational complexity.

[27]  Noam Nisan,et al.  Hardness vs Randomness , 1994, J. Comput. Syst. Sci..

[28]  Sébastien Tavenas,et al.  Improved bounds for reduction to depth 4 and depth 3 , 2013, Inf. Comput..

[29]  Nitin Saxena,et al.  Diagonal Circuit Identity Testing and Lower Bounds , 2008, ICALP.

[30]  Neeraj Kayal An exponential lower bound for the sum of powers of bounded degree polynomials , 2012, Electron. Colloquium Comput. Complex..

[31]  Russell Impagliazzo,et al.  Derandomizing Polynomial Identity Tests Means Proving Circuit Lower Bounds , 2003, STOC '03.

[32]  Pascal Koiran,et al.  Arithmetic circuits: The chasm at depth four gets wider , 2010, Theor. Comput. Sci..

[33]  Amir Shpilka,et al.  Quasipolynomial-Time Identity Testing of Non-commutative and Read-Once Oblivious Algebraic Branching Programs , 2013, 2013 IEEE 54th Annual Symposium on Foundations of Computer Science.

[34]  Nutan Limaye,et al.  Lower bounds for depth 4 formulas computing iterated matrix multiplication , 2014, STOC.

[35]  Jacob T. Schwartz,et al.  Fast Probabilistic Algorithms for Verification of Polynomial Identities , 1980, J. ACM.

[36]  Leslie G. Valiant,et al.  Fast Parallel Computation of Polynomials Using Few Processors , 1983, SIAM J. Comput..